2012/10/16 by Xiaoping Chen, Zhang, Zhongzhang, Chen, Xiaoping +1
Computer Science · #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Formal Methods in Verification #Machine Learning and Algorithms #Reinforcement Learning in Robotics
paper · pdf · doi:10.48550/arxiv.1210.4912
openalex publication_date 2012/10/16 · openalex created_date 2025/10/24 · openalex updated_date 2026/07/28
Planning in partially observable Markov decision processes (POMDPs) remains a\nchallenging topic in the artificial intelligence community, in spite of recent\nimpressive progress in approximation techniques. Previous research has\nindicated that online planning approaches are promising in handling large-scale\nPOMDP domains efficiently as they make decisions "on demand" instead of\nproactively for the entire state space. We present a Factored Hybrid Heuristic\nOnline Planning (FHHOP) algorithm for large POMDPs. FHHOP gets its power by\ncombining a novel hybrid heuristic search strategy with a recently developed\nfactored state representation. On several benchmark problems, FHHOP\nsubstantially outperformed state-of-the-art online heuristic search approaches\nin terms of both scalability and quality.\n